CFA Institute

Markets were generally buoyant in June and July as participants focused on the positives and largely ignored higher risk-free rates and other phenomena with negative implications for asset prices. The ICE BofA US High Yield Index faced resistance in the low 400-basis-point (bp) option adjusted spread (OAS) range, which is consistent with where resistance has been for much of the past year. But patience may be rewarded. Why? Because certain areas of the market have advanced further than the fundamentals justify. Momentum and fear of missing out (FOMO) seem to have driven price movements in June and July. An expensive market that becomes more expensive is one of the more difficult setups for a fundamental- and valuation-driven approach to positioning. Yet despite a market that is fully priced overall, some attractive individual opportunities remain for those willing to search for them. Six or seven months ago, the US high yield market looked likely to be range-bound over the near term. A rally beyond the low 400 bp range in spread appeared challenging. Even though the market broke through 420 bps several times over the past six months or so, it may signal an overextended market rather than one transitioning to a new tighter-spread reality. ICE BofA US High Yield Spread (bps) Source: ICE/Bloomberg There are plenty of signs of late-cycle dynamics. The increased cost of capital over the past 18 months or so has yet to be felt by much of the market. Price action in response to the artificial intelligence (AI) craze has drawn comparisons to the late 1990s tech bubble, and some have argued it maybe years until it peaks. The current market environment is likely an echo of the speculation-driven bubble of 2021 when cryptocurrencies, non-fungible tokens (NFTs), meme stocks, and special purpose acquisition companies (SPACs) were all the rage. AMC, Bed Bath & Beyond, and other stocks had spectacular short-term run-ups well into 2022. It is a bad sign when the main market driver looks like a bubble and the rationale for investing in it is the longevity of the dot-com bubble. After all, that bubble was so detached from reality that the NASDAQ fell by 80% from peak to trough and the US Federal Reserve cut its policy rate by 4.25% on a net basis. While hawkish central bank signals have hurt fixed-income markets this summer, a higher-for-longer regime benefits floating rate securities, including leveraged loans and rate re-set preferred shares. The market has been pricing in higher long-term rates than the Federal Open Market Committee (FOMC) dot plot for months now, but June’s updated forecasts showed relatively significant movement among the voters: Seven of 17 respondents projected a long-term policy rate over 2.5%. In March, only four projected as much, and a year ago only two. And these projections may still be well behind the curve even if they represent a slow acknowledgement of reality. Several areas of the credit markets are showing cracks and structural problems. With many mortgage maturities coming due in the next couple of years, commercial real estate is a particular concern. While this is hardly news to the market, the impact has not been fully appreciated. In leveraged finance, the lack of collateralized loan obligation (CLO) issuance could push more issuers to the high-yield market, increasing the pricing power for investors and the cost of capital for issuers. Now may be a great time to stockpile excess capital to tactically deploy in the coming months if the opportunity set improves. If you liked this post, don’t forget to subscribe to Enterprising Investor. All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer. Image credit: ©Getty Images / Koh Sze Kiat Professional Learning for CFA Institute Members CFA Institute members are empowered to self-determine and self-report professional learning (PL) credits earned, including content on Enterprising Investor. Members can record credits easily using their online PL tracker. source

What I Learned about Investing from Darwin. 2023. Pulak Prasad. Columbia University Press. Investment professionals know that there is no substitute for hours of in-depth textbook study combined with an equal helping of hands-on experience. Self-taught investors, however, can develop significant knowledge and skillsets for their own investing success even without the formal rigor of a professional designation or related university degree. A third group of investors, less inclined to investment theory and practice, may stop at foundational concepts such as risk and return, the benefits of compounding, and the impact of taxes. These three groups are well served by, respectively, high-priced textbooks, detailed investment guides, and retirement planning guides. Star asset manager and founder of Nalanda Capital, a Singapore-based firm, Pulak Prasad has written a timely and practical guide for the middle group, but the book is also a potent reminder to investment professionals that all the technical skills in the world are no substitute for good perspective and strategy. Singapore-based Prasad treads the well-worn path of previous (and perhaps better known in North America) star investors such as Peter Lynch, whose classic guide One Up on Wall Street directed readers to invest in companies they know — in particular, those with abundant compound growth potential. Prasad leverages Lynch’s well-supported wisdom with examples from his India-focused fund but with far greater attention to investment theory and analytical techniques. This level of detail may overwhelm investors who lack a strong grounding in theory and practice, but it is essential to Prasad’s claim that too many professional analysts rely on a false precision that provides answers unrelated to the fundamental question, “Is this company a good long-term investment?” Prasad does not reject the analytical tools but, rather, rejects their unbridled use as hindering analysts’ ability to identify companies that provide superior compound growth and downside protection. He thereby provides an indispensable reminder to chronically underperforming active managers. Prasad does not shy away from detailed commentary on analytical techniques, but he uses a folksy style like Warren Buffett’s to relate each point to real-world examples, often from his own portfolio at Nalanda Capital. Doing so helps the narrative flow, which is much better than in many textbooks — another reason for investment professionals to pick up the book. Prasad highlights his points through well-chosen examples from evolutionary biology, including but not limited to works by Charles Darwin. Each chapter begins with a well-chosen quote from Darwin and from Buffett (who is also liberally referenced in the chapter text) and concludes with a summary of the main points. Prasad’s ability to draw parallels between evolutionary theory and investment theory emphasizes the concepts that are most likely to lead to long-term success and market outperformance. For example, in his second chapter, Prasad cites an evolutionary biology experiment conducted in Siberia in which wild foxes were bred for a “tameness” gene that would make them more like domestic dogs than wild foxes. The experiment began in 1959, and by 1963, it had produced a tamer fox. But the genetic modification also produced other pet-like changes in the animal, such as “floppy ears, a piebald colouration, and a shorter snout,” as well as a shorter reproductive cycle. Prasad draws a parallel between the scientists’ focus on a single desirable trait and his own favored investment metric: return on capital employed (ROCE). He explains that ROCE is likely to be associated with other favorable corporate qualities, such as stellar management, exceptional capital allocation, strong competitive advantage, and capacity to innovate and grow a company. By choosing the primary metric with the most explanatory power, the associated secondary metrics (floppy ears or stellar management) are likely to be attractive. Most analysts are misguided in their use of earnings before interest and taxes (EBIT) or its related measure EBITDA (which includes depreciation and amortization) because those measures can obscure other financial issues. Prasad’s focus on ROCE is an initial screen around which, in the following chapters, he methodically builds his case with additional financial and evolutionary theory, illustrating each with colorful examples. By the book’s conclusion, Prasad has reminded us that the detailed knowledge and refined techniques we acquire through study are not an end in themselves but a means to an end. His perspective is one that draws on experience and demonstrated success and one that investors would do well to emulate. It is also a perspective that may become more valuable in the future as algorithms and artificial intelligence are used to attain financial ends. (More and faster spreadsheets will not help if they do not focus on the best metrics.) The book is clearly written and well edited, with only occasional small missteps. Examples include Prasad’s claim of a zero percent return for an investment that goes bankrupt (that would be a minus 100% rather than a 0% return) and his awkward attempt at humor in suggesting that younger readers may not know what a bookshop is. Also, some of Prasad’s advice seems to lack context. For example, he “detest[s] any debt” on company balance sheets, but public companies with no debt (or even with less debt than they can bear) and without dual class voting structures may be prime candidates for leveraged buyouts. This strategy is a fine potential exit for many active managers but one seemingly at odds with the author’s “buy and hold forever” strategy. These quibbles, however, are small. For amateur and professional investors alike, the book reframes the quest for long-term investment success from a focus on the tools we have to a focus on the outcomes we seek. If you liked this post, don’t forget to subscribe to the Enterprising Investor. All posts are the opinion of the author(s). As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer. Professional Learning for CFA Institute Members CFA Institute members are empowered to self-determine and self-report professional learning (PL) credits earned, including content on Enterprising Investor. Members can

Can stock investors benefit from corporate bond market data? Yes. In fact, equity portfolios constructed using bond momentum signals may outperform their traditional equity price momentum counterparts. But as our study demonstrates, the signal design is critical. Momentum Spillover The momentum spillover effect describes the signal that a company’s bond momentum sends about its future stock returns and is attributed to information asymmetry in the financial markets. There are several reasons why bond market data might have unique insights for equity investors: Institutional investors with advanced expertise and access to more and better data dominate the bond markets relative to their equity counterparts. This may give the bond markets an informational advantage. Since bonds have more predictable future cash flows, their prices may better reflect their fundamental value. Low liquidity and high transaction costs may insulate bond markets from speculation and short-term volatility. Bond Momentum Design Harvesting the momentum spillover premium requires an appropriately designed bond momentum signal. Unlike stock momentum, bond momentum has no single definition. According to the academic literature, bond momentum signals take three forms: Total Return Bond Momentum reflects the aggregated trailing total return of all of a company’s outstanding bonds. Excess Return Bond Momentum describes the difference between the bond total return and duration matched risk-free bond total return. Spread Change Bond Momentum is the negative value of the spread change. In “Momentum in Corporate Bond Returns,” Gergana Jostova et al., examine Total Return Bond Momentum and identify a strong momentum effect in non-investment-grade bonds. But ranking stocks based on bond total return, or interest rate and spread return, may be ill-advised since the former is a systematic factor driven by sovereign interest rate dynamics. As a result, the interest rate exposure of a company’s debt can significantly influence Total Return Bond Momentum. That is why we focus here on Spread Change Bond Momentum and Excess Return Bond Momentum. Applying Bond Momentum to an Equity Portfolio Our bond dataset is based on the Russell 1000 stock universe and starts in 2003, shortly after the launch of the Trade Reporting and Compliance Engine (TRACE) fixed-income database. We mapped corporate bond securities to their stocks using a common company ID. As of December 2022, about 60% of Russell 1000 firms representing 86% of the index’s total market cap have bond data coverage. We computed market-value-weighted excess bond returns and spread changes for all debt-issuing companies with a trailing three-month lookback window and built factor-mimicking portfolios by sorting stocks into quintiles (Q1 to Q5) based on their bond momentum scores. The first chart presents the performance summary of equally weighted and market-cap weighted Q1 to Q5 factor portfolios, along with a Carhart momentum factor portfolio for comparison purposes .  Both bond momentum signals outperformed traditional equity momentum on an equal- and market-cap-weighted basis and had higher information ratios. Furthermore, Spread Change Bond Momentum eclipsed Excess Return Bond Momentum with higher Q1 annualized returns and Q1 to Q5 return spreads. Hypothetical Bond Momentum Portfolio Performance Summary(Russell 1000, 2003 to 2022) Portfolio Excess Return Bond Momentum Spread Change Bond Momentum Equity Momentum Annualized Return Excess Return Information Ratio Annualized Return Excess Return Information Ratio Annualized Return Excess Return Information Ratio Equally Weighted Portfolio Q1 12.2% 1.9% 0.34 12.9% 2.7% 0.41 11.5% 1.3% 0.24 Q2 12.5% 2.3% 0.44 12.6% 2.4% 0.47 11.3% 1.1% 0.28 Q3 12.6% 2.4% 0.47 12.1% 1.9% 0.40 12.0% 1.7% 0.36 Q4 11.3% 1.1% 0.25 11.1% 0.9% 0.23 11.4% 1.2% 0.25 Q5 11.1% 0.9% 0.20 10.9% 0.7% 0.19 12.9% 2.7% 0.29 Q1–Q5 1.1% – – 2.0% – – –1.4% – – Market Cap Weighted Portfolio Q1 10.0% –0.2% 0.04 10.5% 0.3% 0.10 9.3% -0.9% -0.11 Q2 10.9% 0.7% 0.17 11.4% 1.2% 0.29 11.3% 1.1% 0.26 Q3 10.6% 0.4% 0.11 10.7% 0.5% 0.11 10.7% 0.5% 0.14 Q4 10.1% –0.1% –0.02 9.4% –0.8% –0.13 9.3% -0.9% -0.12 Q5 8.8% –1.4% –0.24 7.6% –2.6% –0.36 10.5% 0.3% 0.13 Q1–Q5 1.2% – – 1.9% – – –1.2% – – Source: Northern Trust Quant Equity Research, FTSE Russell, FactSet, Russell 1000The data contained herein does not represent the results of an actual investment portfolio but reflects the hypothetical historical performance. Past Performance is not indicative of future results. Analysis That Spread Change Bond Momentum outperforms Excess Bond Momentum is no coincidence. There are some fundamental explanations for this outcome. Using basic bond math, we decompose bond excess return into spread carry return and spread price return in Equations 1 to 6 below. Spread carry return is a function of spread level while spread price return is driven by spread change. Spread change is the only component that directly captures company-specific market sentiment.  We also applied Fama–Macbeth regressions to further evaluate the two bond momentum signals. Specifically, we ran cross-sectional regressions each month using one-month forward stock returns as independent variables and common stock factors plus bond momentum as dependent variables. The model outputs are presented in the following table. Stock Return and Bond Momentum Factors: Cross-Sectional Analysis, 2003 to 2022   Model 1 Model 2 Model 3 Model 4 Intercept 0.0103 [3.46] 0.0103 [3.44] 0.0106 [3.56] 0.0105 [3.52] Market 0.0024 [1.49] 0.0024 [1.47] 0.0024 [1.45] 0.0024 [1.46] Size 0.0006 [1.59] 0.0006 [1.55] 0.0006 [1.70] 0.0007 [1.85] Value –0.0004 [-0.53] –0.0004 [-0.48] –0.0004 [-0.49] –0.0004 [-0.50] ROE 0.0001 [0.04] 0.0002 [0.06] 0.0001 [0.02] –0.0001 [-0.02] Low Vol 0.0133 [1.55] 0.0126 [1.49] 0.0122 [1.46] 0.0122 [1.45] Momentum 0.0034 [0.85] 0.0029 [0.75] 0.0026 [0.67] 0.0028 [0.71] Excess Return Bond Momentum   0.0357 [1.71]   –0.0072 [-0.25] Spread Change Bond Momentum     0.1957 [2.54] 0.2209 [2.10] R^2 0.1347 0.1382 0.1381 0.1403 Sources: Northern Trust Quant Equity Research, FTSE Russell, FactSet, Russell 3000The data contained herein does not represent the results of an actual investment portfolio but reflects the hypothetical historical performance. Past Performance is not indicative of future results Model 1 is a baseline Fama–French three-factor model plus return on equity (ROE), Low Volatility, and Momentum. Model 2 expands on Model 1 by adding Excess Return Bond Momentum as an independent variable. Model 3 uses Spread Change Bond Momentum as the additional variable, while Model 4 includes

Investment advisors may be overestimating the risk of equities for longer-term investors. We analyzed stock market returns for 15 different countries from 1870 to 2020 and found that optimal equity allocations increase for longer investment horizons. Optimization models that use one-year returns generally ignore the historical serial dependence in returns, so naturally they may over-estimate the risk of equities for longer-term investors, and this is especially true for investors who are more risk averse and concerned with inflation risk. In our previous blog post, we reviewed evidence from our recent paper that returns for asset classes do not evolve completely randomly over time. In fact, some form of serial dependence is present in a variety of asset classes.  While there have been notable differences in the optimal equity allocation across countries, there is significant evidence that investors with longer investment horizons would have been better served with higher allocations to equities historically. It is of course impossible to know how these relations will evolve in the future. However, investment professionals should be aware of these findings when determining the appropriate risk level for a client. Determining Optimal Portfolios Optimal portfolio allocations are determined using a utility function. Utility-based models can be more comprehensive and relevant than defining investor preferences using more common optimization metrics, such as variance. More specifically, optimal asset class weights are determined that maximize the expected utility assuming Constant Relative Risk Aversion (CRRA), as noted in equation 1. CRRA is a power utility function, which is broadly used in academic literature.  Equation 1. U(w) = w-y The analysis assumes varying levels of risk aversion (y), where some initial amount of wealth (i.e., $100) is assumed to grow for some period (i.e., typically one to 10 years, in one-year increments). More conservative investors with higher levels of risk aversion would correspond to investors with lower levels of risk tolerance. No additional cash flows are assumed in the analysis. Data for the optimizations is obtained from the Jordà-Schularick-Taylor (JST) Macrohistory Database. The JST dataset includes data on 48 variables, including real and nominal returns for 18 countries from 1870 to 2020. Historical return data for Ireland and Canada is not available, and Germany is excluded given the relative extreme returns in the 1920s and the gap in returns in the 1940s. This limits the analysis to 15 countries: Australia (AUS), Belgium (BEL), Switzerland (CHE), Denmark (DNK), Spain (ESP), Finland (FIN), France (FRA), UK (GBR), Italy (ITA), Japan (JPN), Netherlands (NLD), Norway (NOR), Portugal (PRT), Sweden (SWE), and United States (USA).  Four time-series variables are included in the analysis: inflation rates, bill rates, bond returns, and equity returns, where the optimal allocation between bills, bonds, and equities is determined by maximizing certainty-equivalent wealth using Equation 1. Three different risk aversion levels are assumed: low, mid, and high, which correspond to risk aversion levels of 8.0, 2.0, and 0.5, respectively. These, in turn, correspond approximately to equity allocations of 20%, 50%, and 80%, assuming a one-year investment period and ignoring inflation. The actual resulting allocation varies materially by country. Any year of hyperinflation, when inflation exceeds 50%, is excluded. Exhibit 1 includes the optimal equity allocation for each of the 15 countries for five different investment periods: one, five, 15, and 20 years, assuming a moderate risk tolerance level (y=2) where the optimizations are based on the growth of either nominal wealth or real wealth, using the actual historical sequence of returns or returns that are randomly selected (i.e., bootstrapped) from the historical values, assuming 1,000 trials. The bootstrapping analysis would capture any skewness or kurtosis present in the historical return distribution because it is based on the same returns, but bootstrapping effectively assumes returns are independent and identically distributed (IID), consistent with common optimization routines like mean-variance optimization (MVO). Exhibit 1. Optimal Equity Allocations for a Moderate Risk Aversion Level by Country and Investment Period: 1870-2020 Important Takeaways There are several important takeaways from these results. First, there are considerable differences in the historical optimal equity allocations across countries, even when focusing on the same time horizon (one-year returns). For example, the equity allocations range from 16% equities (for Portugal) to 70% (for the United Kingdom) when considering nominal, actual historical returns.  Second, the average equity allocation for the one-year period across all 15 countries is approximately 50%, regardless of whether wealth is defined in nominal or real terms. Third, and perhaps most notably, while the equity allocations for the optimizations using actual historical return sequences increase over longer investment optimizations, there is no change in optimal allocations for the bootstrapped returns. The equity allocations for the nominal wealth optimizations increase to approximately 70% at 20 years, and equity allocations for the real wealth optimizations increase to approximately 80% at 20 years, which represent annual slopes of 1.3% and 1.5%, respectively. In contrast, the equity allocations for the boostrapped optimizations are effectively constant (i.e., zero). This finding is worth repeating: the optimal allocation to equities is different using actual historical return data (which have nonzero autocorrelation) than in the bootstrapped simulation where returns are truly IID. Exhibit 2 includes the average allocations to equities across the 15 countries for the three different risk aversion levels when focused on nominal and real wealth and on whether the actual historical sequence of returns are used or if they are bootstrapped. Note, the average values in Exhibit 1 (for the one, five, 10, 15, and 20 year periods) are effectively reflected in the results in the next exhibit for the respective test. Exhibit 2. Optimal Equity Allocation by Risk Tolerance Level and Investment Period (Years) Again, we see that optimal equity allocations tend to increase for longer investment periods using actual historical return sequences, but the bootstrapped optimal allocations are effectively constant across investment horizons. The impact of investment horizon using the actual sequence of returns is especially notable for the most risk averse investors. For example, the optimal equity allocation for an investor with a high-risk aversion level focused on nominal

Factors are the primary market drivers of asset-class returns. In the equity realm, only a limited set of rewarded factors are backed by academic consensus: Value, Size, Momentum, Low Volatility, High Profitability, and Low Investment. These factors compensate investors for the additional risk exposure they create in bad times. Hence, factor strategies are appealing to investors because they provide exposure to rewarded risk factors in addition to market risk and can be a source of superior risk-adjusted performance over the long term compared with cap-weighted benchmarks. The year 2022 was a memorable one for investors, but for not altogether positive reasons. One bright spot, however, was the relative outperformance of equity risk factors versus other popular equity investing styles. While the financial media has attributed recent strong factor performance almost entirely to the Value factor, the resurgence of factor performance was in fact much broader. Factor Performance’s Comeback Was Broad Based Here “factor performance” refers to the performance of long/short factor portfolios that go long a subset of stocks with the strongest positive exposure to a given factor and short a subset of stocks with the strongest negative exposure to the same factor. Indeed, in the United States, almost all factors had positive performance in 2022, with an average return of 6.9%, which is in line with their long-term average, as illustrated in the chart below. Momentum, Low Investment, and Value factors beat their long-term average, though not their best 5% annual rolling returns. The Low Volatility and Size factors also had positive performance albeit below their long-term average. High Profitability was an outlier, posting the only negative performance. Indeed, the factor fared so poorly, it eclipsed its worst 5% rolling return between 31 December 1974 and 31 December 2021. US Factor Performance in 2022 US Factors Size Value Mom Low Vol High Pro Low Inv 6-F EW 2022 3.5% 8.4% 19.9% 4.3% -10.1% 15.4% 6.9% Avg. RollingAnnual Return 8.8% -1.7% 3.9% 8.5% 3.8% 4.1% 4.1% Worst 5% Rolling Return -22.0% -20.5% -20.9% -17.4% -9.1% -9.2% -3.9% Best 5%Rolling Return 53.8% 14.4% 27.9% 36.9% 22.5% 21.3% 18.7% Size, Value, Momentum, Low Volatility, High Profitability, and Low Investment are Scientific Beta long/short market beta neutralized factors used in seven-factor regressions. The worst/best 5% one-year return corresponds to the 5th and 95th percentile of one-year rolling return with a weekly step over the period from 31 December 1974 to 31 December 2021. The results in the chart above contradict two popular media narratives: that the factor performance story is solely a Value story and that any highly profitable company will outperform in a rising rate environment. The Factor Story Has Been a Sector Story Which sectors drove factor performance in 2022? The energy sector played an outsized role. It outperformed its broad cap-weighted counterpart by 84.5% and, as the exhibit below illustrates, helped drive Value, Momentum, and Low Investment factor performance and negatively impacted Low Volatility and High Profitability. Sector Performance Attribution: US Factors, 2022 The graph represents the sector performance attribution of each L/S rewarded factor in 2022 without accounting for market beta adjustment. For international equities and global equities, the story is largely consistent with the US market. Factor Performance through a Macro Lens While macro factors are not the primary drivers of equity performance, they can have significant influence on factor behavior in certain environments. In examining how the macro environment influences factor performance, we use a macro framework developed by Noël Amenc, Mikheil Esakia, Felix Goltz, and Ben Luyten. Our four macro variables, shown in the chart below, are short rates (three-month Treasury bills); term spread (10-year minus 1-year Treasuries); default spread (Baa minus Aaa Corporate Bonds); and breakeven inflation (10-year break-even inflation). For each macro variable, we build a long/short macro portfolio composed of stocks with the strongest and weakest sensitivity to macro innovations (surprises). We go long stocks with the highest sensitivity to weekly macro innovations and short stocks with the lowest sensitivity to weekly macro innovations. In 2022, macro factors explained much of the variability of some US equity factors. For instance, term spread, credit spread, and breakeven inflation factors, respectively, explained 27%, 33.7%, and 45.3% of the Value factor’s variability over the period. Breakeven inflation was one of the strongest macro factors since it explained a large part of the return variability of Value, High Profitability, and Momentum. No macro factor had a real impact on the variability of the Momentum factor. Percentage of 2022 US Equity Factor Performance Explained by Macro Factors US 2022R-Squared Size Value Momentum LowVolatility HighProfitability LowInvestment Short Rate 6.1% 0.4% 0.6% 46.7% 8.0% 1.0% Term Spread 8.6% 27.0% 1.2% 36.3% 36.5% 11.7% Credit Spread 11.4% 33.7% 5.3% 20.5% 47.1% 22.4% Breakeven Inflation 12.5% 45.3% 7.1% 19.6% 67.0% 29.7% The results above are a contrast to the longer-term impact of macro factors on equity factors, depicted in the following chart. While macro factors do not have the most significant impact over the longer term, given the transition to a more normalized interest rate environment, they do exert a more pronounced effect on 2022 factor performance. This is consistent with academic findings. Indeed, factor risk premia short-term variations are linked to the business cycle or macroeconomic conditions. Percentage of US Equity Factor Longer-Term Performance Explained by Macro Factors US Long-TermR-Squared Size Value Momentum LowVolatility HighProfitability LowInvestment Short Rate 0.9% 5.9% 6.0% 29.4% 1.2% 14.5% Term Spread 1.9% 1.2% 0.0% 14.9% 3.7% 0.8% Credit Spread 4.7% 0.3% 0.0% 21.7% 0.0% 7.1% Expected Inflation 0.4% 3.2% 0.2% 4.9% 10.3% 0.8% How did macro factors affect equity factors? The chart below shows Value and Low Investment had positive sensitivity and High Profitability and Low Volatility negative sensitivity to breakeven inflation. Similarly, Value and Low Investment had negative sensitivity and Low Volatility and High Profitability positive sensitivity to the credit spread factor. 2022 US Equity Factor Sensitivities to Macro Factors US 2022Betas Size Value Momentum LowVolatility HighProfitability LowInvestment Short Rate 0.22 0.05 -0.04 -1.11 -0.25 -0.08 Term Spread 0.16 0.33 0.07 -0.62 -0.35 0.23 Credit Spread -0.33 -0.65 -0.34 0.83

“I think, therefore I am.” —René Descartes Investing isn’t a test of who’s right; it’s a test of who updates best. In that scenario, success doesn’t go to those with perfect predictions, it goes to those who adapt their views as the world changes. In markets shaped by noise, bias, and incomplete information, the edge belongs not to the boldest but to the most calibrated. In a world of uncertainty and shifting narratives, this post proposes a new mental model for investing: Bayesian edge investing (BEI) — a dynamic framework that replaces static rationality with probabilistic reasoning, belief-calibrated confidence, and adaptive diversification. This approach is an extension of Bayesian thinking — the practice of updating one’s beliefs as new evidence emerges. For investors, this means treating ideas not as fixed predictions but as evolving hypotheses — adjusting confidence levels over time as new, informative data become available. Unlike modern portfolio theory (MPT), which assumes equilibrium and perfect foresight, BEI is built for a world in flux, one that demands constant recalibration rather than static optimization. A confession: Much of what I’ve explored in this post remains a work in progress in my own investment practice. Judgment Over Analysis Financial models are teachable. Judgment is not. Most frameworks today are centered on mean-variance optimization, assuming investors are rational, and markets are efficient. But the reality is messier: markets are often irrational, and investor beliefs evolve. At its core, investing is a game of decisions under uncertainty, not just numbers on a spreadsheet. To consistently outperform, investors must confront irrationality, navigate evolving truths, and react with rational conviction — a much harder task. That means shifting from deterministic models to belief-weighted, evidence-updated frameworks that recognize markets as adaptive systems, not static puzzles. Calibrated, Not Certain In investing, being rational isn’t about being certain. It’s about being calibrated. It’s about recognizing irrationality and then responding with discipline, not emotion. But here’s the paradox: both irrationality and rationality are elusive and often indistinguishable in real time. What appears obvious in hindsight is rarely clear in the moment, and this ambiguity fuels the very boom-bust cycles investors try to avoid. BEI reframes rationality as the ability to construct a probability-weighted map of future outcomes and to continuously update beliefs as new information emerges. It is: Bayesian, because beliefs evolve with evidence. Edge-seeking, because alpha lies in misalignments between an investor’s belief and the market’s. Rationality in this framework means acting when your updated model of reality diverges materially from prevailing prices. A Mental Model: Truth ≈ ∫ (Fact × Wisdom) d(Reality) “Truth” based on facts and wisdom leads to “Reality.” “Facts” are objective but “Truth” is conditional. It emerges from how much information is available and how well you interpret it. Let’s reframe how we perceive “Truth” in markets. It is a function of: Facts — objective data. Wisdom — Interpretive ability, including judgement and context. Together, facts and wisdom determine how close our perception of truth aligns with reality. Like an asymptote, we approach reality but never fully capture it. The goal is to move further along the truth curve than other market participants. Figure 1 illustrates this relationship. As both relevant data (facts) and interpretive wisdom increase, our understanding (truth) moves progressively closer to reality – asymptotically approaching it, but never fully capturing it in advance. Figure 1. This mental model reframes rationality as the pursuit of superior probabilistic judgment. Not certainty. It’s not about having the answer, but about having a more informed, better-calibrated answer than the market. In other words, aiming to be further along the truth curve (reality). From Bias to Bayes Cognitive biases like loss aversion, confirmation bias, and anchoring cloud decisions. To combat these biases, Bayesian thinking begins with a hypothesis and updates belief strength in proportion to the diagnostic power of new information. Not every data point deserves equal weight. The disciplined investor must ask: How likely is this information under competing hypotheses? How much weight should it carry in updating my conviction? This is dynamic conviction-building rationality in motion. A Biotech Case Study The principles of BEI come into sharper focus when applied to a real-life decision-making exercise. Imagine a mid-cap biotech firm developing a breakthrough therapy. You initially place the probability of success at 25%. Then the company announces positive and statistically significant Phase II trial results — a meaningful signal that warrants a reassessment of the initial belief. Bayesian Update: P(Positive Result | Success) = 0.7 P(Positive Result | Failure) = 0.3 P(Success) = 0.25 P(Failure) = 0.75 Bayesian Update: P(Success | Positive Trial) = [P(Positive Trial | Success) × P(Success)] / {[P(Positive Trial | Success) × P(Success)] + [P(Positive Trial | Failure) × P(Failure)]} = (0.7 × 0.25) / [(0.7 × 0.25) + (0.3 × 0.75)] = 0.175 / 0.4 = 0.4375 → 43.75% This increases confidence in the trial’s success from 25% to 43.75%. Now embed this in a Weighted Evidence Framework: A single data point can meaningfully shift conviction, position sizing, or risk exposure. The process is structured, repeatable, and insulated from emotion. Interpretation: Understanding what the market implicitly believes can reveal powerful opportunities. In the example discussed, if the current price of $50 reflects only existing cash flows and an additional $30 of value is estimated with 57% confidence, the gap suggests a potential analytical edge — one that could justify a high-conviction position. Turning Confidence into Allocation Traditional diversification assumes perfect calibration and constant correlations. BEI proposes a different principle: allocate based on your edge. This framework constructs portfolios based on two factors: an investor’s dynamically updated confidence level in a thesis and the investor’s assessment of market irrationality, or perceived mispricing. Unlike traditional models that theoretically push all investors toward a similar optimal portfolio, this approach generates a personalized investment universe, inherently discouraging “me-too” trades and aligning capital with an investor’s unique insight. This framework positions ideas across two axes: conviction and the magnitude of mispricing: Why this works: Depth over breadth — Focus capital where you

Investing in the Era of Climate Change. 2022. Bruce Usher. Columbia University Press. The scientific consensus is that climate change is real, occurring now, and potentially catastrophic. As a result, most countries have committed to reductions in greenhouse gas emissions with the aim of “net zero” emissions by the middle of the 21st century. To achieve the reductions, innovation and investment are needed on a large scale. Bruce Usher of Columbia Business School approaches the issue from the perspective of the investor, and in Investing in the Era of Climate Change, he identifies both what the implications of climate change are for the investment community and how investment capital allows us “to save us from ourselves.” The role of investors, he says, is no less than “financing the world’s future.” Early in the book, Usher gives an account of technological developments that can mitigate the effects of climate change — renewable power, electric vehicles, battery storage, green hydrogen, and carbon removal. This discussion serves as a valuable introduction to later sections that deal with the implications of such climate solutions for the investment community. One section identifies the alternative strategies that the investor can use: Risk Mitigation Divestment Environmental, Social, and Governance (ESG) Investing Thematic Impact Investing (to finance businesses that address a specific environmental or social challenge, such as climate change) Impact First Investing (in which investors focus on solving social and environmental problems and are willing to accept a below-market financial return in exchange for greater impact) Each of these strategies is suitable for a particular kind of investor. University endowments may opt for Divestment, large fund managers for ESG, specialist fund managers for Thematic Impact Investing, and philanthropists for Impact First Investing. Some approaches help to control risks; others (according to Usher) can improve returns. Asserting that “all investors should understand the opportunities and risks of investing in real assets that offer climate solutions,” the author then looks at both financial and real assets. Real assets include renewable energy projects, real estate, and forestry and agriculture. His analysis examines the valuation issues relevant to large-scale renewables projects, along with insights into government incentives and prospective returns (internal rates of return of 6%–8% for solar and wind projects and potentially more return for higher risk investments in battery energy storage systems). The discussion of real estate is brief but includes such considerations as the risks from flooding and wildfires as well as the benefits of energy upgrades — the Empire State Building is an interesting example. The importance of carbon markets is illustrated by the chapter on forestry and agriculture. The author’s analysis of financial assets includes chapters on venture capital, private equity, public equity, equity funds, and fixed income. We are given interesting examples of successful and unsuccessful investments, along with the following approaches to assessing investments in the era of climate change: Is a company minimizing risk by reducing its emissions, both direct and indirect? What would be the impact of a price on carbon? Is the company an incumbent in an industry or a disruptor? If a disruptor, how likely is it to succeed? The chapter on equity funds identifies many types of currently available climate-focused funds and exchange-traded funds (ETFs). The analysis covers the differences among low-carbon funds, fossil-fuel-free funds, and climate transition funds. The author notes that some of these funds are particularly large and successful: “BlackRock’s Carbon Transition Readiness ETF pulled in $1.3 billion on its first day of trading, making it the biggest launch in the ETF industry’s three-decade history.” A successful fund launch is one example of how investing in climate solutions has become mainstream. So too is the establishment of such bodies as the Glasgow Financial Alliance for Net Zero — “a global coalition of 450 financial firms managing assets of more than $130 trillion that are committed to reducing greenhouse gas emissions to zero.” The author believes that the fixed-income markets will be the most important for the funding of climate solutions. Part of the reason is their scale, and part is because many projects, with steady cash flows over long periods of time, lend themselves to debt financing. An important area is that of “green bonds,” the market for which is described as “red hot.” In 2021, $500 billion of green bonds were issued. Other innovations in fixed-income investing include the securitization of solar leases and loans. Several times throughout this book, we read estimates of the costs of necessary climate solutions. The various numbers can be confusing, but all are broadly consistent with a Boston Consulting Group estimate of what is required: $3 trillion to $5 trillion per year. This enormous level of investment is a huge step up from where we are today (spending of circa $600 billion a year, according to Usher). The investment is necessary, however, especially because other possible responses to climate change can be convincingly rejected. (These alternatives include adaptation and the control of population growth.) A welcome aspect is that the general tone of the book is upbeat, with a focus on solutions rather than resorting to despair. At times, however, this approach means glossing over certain risks to climate targets. For example, livestock make a material contribution to greenhouse gases (in the form of methane), but apart from references to the success of Beyond Meat, the author offers us few solutions to the issue of livestock. Similarly, he says little about how to mitigate emissions caused by the production of cement. Furthermore, although he does write that “perhaps the greatest challenge to reaching net zero is the inability by countries to cooperate,” he says little about how dependent we are on fragile global supply chains for solutions, such as battery storage systems. The author makes clear, however, that his goal is not to describe every possible solution to the climate crisis but to focus on the implications of climate change for investors. Investing in the Era of Climate Change draws from a wide variety of sources and is

I recently sat down with Jason Hsu, founder of Rayliant Global Advisors and chief economist of East West Bank, to discuss the evolution of factor investing, the challenges facing the asset management industry, and the opportunities offered by modern technologies and approaches. This interview is part of the Conversations with Frank Fabozzi, CFA series, sponsored by the Research and Policy Center. The series aims to bring leading experts in finance and economics into dialogue to explore critical issues shaping the industry’s future. Hsu is a recognized leader in quantitative asset management and co-founder of Research Affiliates. You can register for my upcoming conversation with Lori Heinel, CFA, EVP and global chief investment officer at State Street Global Advisors here. Hsu’s reflections in this session underscore the shifts in investment paradigms, the growing pressures on asset managers to differentiate themselves, and the critical role of governance, innovation, and long-term thinking in navigating an increasingly competitive and complex environment. Expanding the Factor Universe Hsu begins by tracing the origins and evolution of factor-based strategies. Initially rooted in academic finance, these strategies have become staples in institutional and retail investing. Traditional factors, such as value, momentum, and size, continue to play a significant role, but Hsu highlights a growing appetite for expanding the factor universe. Today, asset managers are increasingly incorporating macroeconomic signals, such as interest rate changes or inflation dynamics, alongside behavioral factors driven by market psychology. This broadening of the factor toolkit reflects both a response to market commoditization and a recognition that traditional factors, while still valuable, cannot alone address the complexities of modern financial markets. One of Hsu’s key points is the importance of grounding factor-based strategies in clear economic rationale. He warns against over-reliance on historical data or data-mining approaches that lack theoretical justification. While backtesting can yield impressive results, strategies derived without a solid understanding of their underlying drivers risk failing in real-world conditions. Hsu argues that robust factor strategies should be built upon empirical evidence and an intuitive understanding of how and why certain relationships persist across different market environments. This combination ensures that factors remain relevant and effective even as market dynamics evolve. The commoditization of basic factor strategies is a central theme of Hsu’s discussion. As quantitative tools and techniques have become more accessible, the barriers to implementing traditional factor models have diminished. This has led to declining fees and heightened competition among asset managers, pressuring firms to differentiate themselves through innovation. Hsu notes that differentiation often entails exploring new or custom factors, but it also requires maintaining transparency and aligning with client expectations. Firms must balance pushing the boundaries of innovation and delivering strategies that investors can understand and trust. Structural Challenges in Asset Management Hsu also addresses the structural challenges within the asset management industry, particularly those related to governance and incentives. He critiques the pervasive short-termism that dominates many investment decisions, arguing that this mindset often misaligns with the long-term goals of institutional and retail investors. The pressure to deliver quarterly results frequently leads to strategies prioritizing immediate performance over sustainable value creation. Hsu advocates for governance structures that reward long-term thinking and encourage asset managers to focus on delivering outcomes that align with their clients’ broader objectives. The role of technology in reshaping asset management is another critical focus of the interview. Hsu acknowledges the transformative potential of machine learning and artificial intelligence in modern portfolio management. These technologies enable asset managers to uncover complex patterns, process vast datasets, and develop more sophisticated models. Hsu cautions against the indiscriminate use of technology, highlighting the risks of overfitting and the lack of interpretability in many machine learning models. In finance, where decisions often have significant consequences, the inability to explain how a model arrived at its conclusions can undermine its practical value. Hsu argues for a balanced approach to integrating machine learning (ML) with traditional financial and economic theory. Rather than replacing established methodologies, ML should complement them by enhancing the understanding of complex relationships and providing new insights. This integration ensures that models remain robust and interpretable, enabling portfolio managers to leverage the strengths of advanced analytics without sacrificing transparency or trust. Rigorous, Data-Driven Approaches to ESG Needed The increasing prominence of environmental, social, and governance (ESG) investing forms another key theme in my conversation with Hsu. He observes that demand for sustainable investment strategies has grown significantly, driven by both institutional mandates and shifting societal expectations. However, incorporating ESG considerations into investment processes presents unique challenges, particularly in quantifying ESG impact and integrating it into traditional portfolio frameworks. Hsu emphasizes the need for rigorous, data-driven approaches to ESG investing to ensure that it goes beyond superficial claims or “greenwashing.” By aligning ESG metrics with broader financial goals, asset managers can develop strategies that are both impactful and economically viable. Diversity within investment teams is another area where Hsu sees significant opportunities for improvement. He argues that fostering intellectual diversity and encouraging collaboration are essential for success in the evolving asset management landscape. Diverse teams bring varied perspectives and approaches to problem-solving, which can enhance creativity and adaptability. In an industry where market conditions and client demands constantly change, the ability to think critically and adapt quickly is invaluable. One of the most compelling aspects of my conversation with Hsu is his discussion of the challenges and opportunities in implementing factor-based strategies in real-world market dynamics. He notes that value and momentum are not static but evolve as markets change. This evolution requires constant re-evaluation and adaptation of strategies to ensure their continued relevance. Hsu highlights the importance of stress-testing factor models under different scenarios to assess their robustness and potential vulnerabilities. Customization is Key Hsu also reflects on the growing role of customization in asset management. As clients demand more tailored solutions, firms must develop strategies that address specific needs and objectives. This customization often involves creating unique factor combinations or integrating non-traditional data sources, such as alternative datasets, to enhance predictive accuracy. By aligning strategies with client-specific

In his keynote speech at the Ben Graham Centre’s 2024 Value Investing Conference in Toronto, Jason Zweig, a veteran columnist for The Wall Street Journal, asked, rhetorically: “What cannot be ETF’d?” Active investors are competing with Mr. Market, a.k.a. passive exchange traded funds, he stated. To generate meaningful alpha, portfolio managers must develop expertise in what cannot be packaged into an exchange traded fund, Zweig advised.   The target universe for active managers is what Zweig called “left tail things” like size, liquidity, marketability, and popularity factors. These are the factors inherent in small-cap companies. The spring is coiled tight in the small-cap space, and we view it as highly favorable for generating alpha. While small-cap companies are ETF’d, passive investing in this group is a sub-optimal strategy for the creation of alpha over the long term. Portfolio managers must develop expertise in this market segment. In the United States, small- and micro-cap stocks have lagged large- and mega-cap stocks for nearly a decade, based on price returns from the Russell 2000 and the S&P 500. Over the same period, the small-cap effect remained intact in the UK, Japan, Europe, and the emerging markets. What accounts for the outlier status of the United States? Institutional allocations have shifted toward private equity and away from public markets. Global private equity AUM is expected to grow to US$8.5 trillion by 2028, and American firms are leading the charge with a CAGR of 11.3%, according to Prequin. Today, fast-growing smaller companies have financing options they did not have previously. They can stay private much longer, living and growing inside the gated community of private equity. Some of these companies may never join the Russell 2000. If they grow to a sufficient size, they may jump directly to the S&P 500 or be sold to another large private equity fund.  Spotlight on Canada In Canada, the small- and micro-cap space has been in a bear market. Active small-cap-focused funds have seen outflows for the past 10 years, M&A activity is tepid, and IPO activity is weak. The total public capital raise for tech in Canada last year was down 88% from 2022 levels and 98% from 2021 levels. This has created a negative feedback loop in Canada of fleeing capital and underperformance in this sector. During the first quarter of 2024, we saw the first glimmers of change with the S&P TSX Small Cap Index (7.9%) outperforming the S&P TSX Composite Index (6.6%). Outlook for North America Market valuations rose in 2023, which should entice some private companies to go public this year or next. Any improvement in IPO and M&A activity would be a positive tailwind for small caps, which are undervalued on both an absolute and relative basis. We see a target-rich environment in small caps. The lack of research and capital has left the field wide open for astute investors. Potential catalysts for a re-rating will be improved balance sheets, increased cash flow metrics, and increased M&A and IPO activity. Tailwinds include the inflection point on rising interest rates, quality companies continuing to compound business value and clean up their balance sheets, accelerating M&A activity to take advantage of discounted valuations, and mean reversion to historical valuations and sentiment levels. The small-cap sector is best approached through an active investment strategy where expertise and a deep understanding of the individual businesses and their risk-and-reward characteristics are necessary for success. Every investor who strives for outperformance must take on potential risks, however, one of which could be periods of painful unpopularity and underperformance like we have endured in the small-cap sector since 2016. As the Norwegian chess master Magnus Carlsen has said, “Not being willing to take risks is an extremely risky strategy.”  If you liked this post, don’t forget to subscribe to the Enterprising Investor. All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer. Image credit: ©Getty Images/Hanis Professional Learning for CFA Institute Members CFA Institute members are empowered to self-determine and self-report professional learning (PL) credits earned, including content on Enterprising Investor. Members can record credits easily using their online PL tracker. source

The Economics of the Stock Market. 2022. Andrew Smithers. Oxford University Press. Judging by the behavior of the stock market, we are living in challenging times for mainstream finance. Under the hypothesis that markets are efficient and investors are rational, neoclassical theory assumes away the problem of financial bubbles and the linkages between equity returns and the rest of the macro variables. After a decade of unconventional monetary policies, massive fiscal deficits, and the return of inflation, however, equity market behavior in recent years has been nothing short of perplexing, leaving most practitioners struggling to understand the vagaries of stock markets. Today, the workhorse neoclassical model requires a thorough review of its assumptions (and conclusions). Now, more than ever, we urgently need a comprehensive alternative. Andrew Smithers attempts to fill in this gap with his latest book, The Economics of the Stock Market, which offers an alternative theory of how stock markets work. The book builds on a small and obscure tradition of growth models, pioneered by Nicholas Kaldor more than 50 years ago, which dealt with distributional issues in a Harrod–Domar-type framework. One of these iterations showed that in a closed economy with two sectors (households and firms) and no government activity, equity valuation multiples are determined solely by macroeconomic variables — crucially, by the equilibrium between aggregate savings and aggregate investment. Kaldor’s framework was quite novel in that stock market valuations integrated seamlessly into the macroeconomy and were responsible for balancing saving and investment, in contrast to the Keynesian and neoclassical traditions in which the equilibrium process works through quantities (unemployment rate) and prices, respectively. Although Kaldor never intended his model to be a framework for understanding stock markets, Smithers draws on this setup to articulate a theoretical alternative. Smithers is also very “Kaldorian” in the way he constructs his framework, for two reasons. First, he is primarily interested in the long-run behavior of the system, or steady-state solutions. Second, he relies on several “stylized facts” about stock markets to inform his assumptions. In particular, four variables have historically been mean-reverting to a constant, and any model should take these into consideration: Equity returns in real terms The shares of profits (after depreciation) and labor in total output The ratio of interest payments to profits The ratio of the value of fixed capital to output (a Leontief-type production function) The first stylized fact has particular relevance to the mechanics of the overall model. For Smithers, equity returns (in real terms) are mean-reverting and tend toward a constant in the long run, at about 6.7% per annum. According to the author, this long-run constant results from capital owners’ risk aversion rather than from the marginal productivity of capital or from households’ consumption decisions. As we shall see, this dynamic has profound implications for determining returns in other asset classes. This novelty is not the only one in Smithers’s framework. His model varies from the neoclassical framework in at least three other ways. First, at the heart of Smithers’s proposal is the firm as a separate entity from households. This distinction is important because firms behave significantly differently from households. For firms, decisions on investment, dividend policy, share issuance, and leverage are made by managers whose motivation (keeping their jobs) differs substantially from that of the neoclassical utility-maximizing consumer. In Smithers’s framework, firms do not seek to maximize profits, because if they did, they would vary their investments with the cost of capital — as in investment models based on the Q ratio. Casual empirical observation appears to confirm this point — as Smithers explains, “Rises in the stock market would be constrained by a growing flood of new issues as share prices rise and their falls would be limited by their absence in weak markets. Smaller fluctuations in the stock market would seem naturally to follow.” In this respect, any model should also consider the contrasting behavior of listed and unlisted companies. According to Smithers, one consequence of more companies being listed is that the corporate sector as a whole becomes less responsive to the cost of equity (Q models). This dynamic occurs because when it comes to investment decisions, management teams’ behavior is constrained by the possibility of a hostile takeover and job loss. In other words, “managements are concerned with the price of their companies’ shares, rather than the overall level of the stock market.” One macroeconomic implication of the absence of a link between valuations and investment is that the stock market plays an important role in economic growth, by preventing fluctuations in the cost of capital from affecting the level of investment — and ultimately output. Second, the returns among asset classes are derived in an independent fashion and are not codetermined. In Smithers’s framework, a firm’s balance sheet is assumed to consist of short-term debt (which can be thought of as very liquid instruments), long-term bonds, and equity. These instruments’ returns are derived independently, and their influences on the system work through different mechanisms. Savings and investment are equated by movements in the short-term interest rate. Corporate leverage is balanced with the preferences of the owners of financial assets through variations in bond yields. Finally, as explained earlier, equity returns are stationary. Consequently, the difference in returns among asset classes — that is, the equity risk premium — is not mean reverting, it has not historically had a stable average, and its level cannot provide any information about future returns for either equities or bonds. For Smithers, the equity risk premium is a residual and bears little relationship to the role it plays in mainstream finance. Finally, for Smithers the cost of capital varies with leverage at the macroeconomic level. This conclusion diametrically opposes the 1958 Miller–Modigliani Theorem (M&M), which states that the value of a firm is independent of its capital structure. According to M&M, a firm’s risk increases with its financial leverage, so the required return on equity increases with it, leaving the overall cost of capital unchanged because debt is